International Journal of

Environmental Research
and Public Health

Article
Krebs von den Lungen-6 as Disease Severity Marker for
COVID-19 Patients: Analytical Verification and Quality
Assessment of the Tosoh AIA-360 Compared to
Lumipulse G600II
Miriana d’Alessandro 1, *,† , Laura Bergantini 1,† , Dalila Cavallaro 1 , Sara Gangi 1 , Paolo Cameli 1 ,
Edoardo Conticini 2 , Siena COVID Unit ‡ , Bruno Frediani 2 , Francesco Dotta 3 and Elena Bargagli 1

1 Respiratory Diseases and Lung Transplantation Unit, Department of Medical and Surgical
Sciences & Neurosciences, University of Siena, 53100 Siena, Italy; laurabergantini@gmail.com (L.B.);
dalila.cavallaro@student.unisi.it (D.C.); sara.gangi@student.unisi.it (S.G.); paolocameli88@gmail.com (P.C.);
bargagli2@unisi.it (E.B.)
2 Rheumatology Unit, Department of Medicine, Surgery and Neurosciences, University of Siena,
53100 Siena, Italy; conticini.edoardo@gmail.com (E.C.); fredianibruno60@gmail.com (B.F.)
3 Diabetes Unit, Department of Medicine, Surgery and Neurosciences, University of Siena and Fondazione
Umberto Di Mario ONLUS-Toscana Life Science Park, 53100 Siena, Italy; francesco.dotta@unisi.it
* Correspondence: dalessandro.miriana@gmail.com; Tel.: +39-0577586713; Fax: +39-0577280744
† These authors equally contributed to the study.





 ‡ Collaborators/Membership of the Siena COVID Unit Name is provided in the Acknowledgments Section.

Citation: d’Alessandro, M.;

Bergantini, L.; Cavallaro, D.; Gangi,
S.; Cameli, P.; Conticini, E.; Siena
COVID Unit; Frediani, B.; Dotta, F.;
Bargagli, E. Krebs von den Lungen-6
as Disease Severity Marker for
COVID-19 Patients: Analytical
Verification and Quality Assessment
of the Tosoh AIA-360 Compared to
Lumipulse G600II. Int. J. Environ. Res.
Public Health 2022, 19, 2176. https://
doi.org/10.3390/ijerph19042176
Academic Editor: Paul B.
Tchounwou
Received: 27 December 2021
Accepted: 8 February 2022
Published: 15 February 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Abstract: Background: Krebs von den Lungen-6 (KL-6) has been proposed as a disease severity
marker of COVID-19. All research articles reported the KL-6 assay detected through Fujirebio
reagents by Lumipulse G600/G1200 instrument. In the present study, KL-6 assay was analysed
through Tosoh AIA-360 and compared with analytical results by Lumipulse G600 in a population of
COVID-19 patients. Materials and methods: Sixty-four patients (median age, IQR 67 (58–76) years), all
hospitalized for COVID-19 interstitial pneumonia at Siena COVID Unit. KL-6 was measured by two
methods, chemiluminescence enzyme immunoassay (CLEIA) and fluorescent enzyme immunoassay
(FEIA) method by Lumipulse G600 II and AIA 360 systems, respectively. Results: KL-6 concentrations
evaluated by Lumipulse G600II were significantly higher in severe than those in non-severe patients
(p < 0.0001) as well as evaluating by AIA360 (p < 0.0001). Receiver operating curve (ROC) curve
analysis showed that KL-6 concentrations, by Lumipuse G600II, distinguished severe from non-severe
COVID-19 patients with an area under the curve (AUC) of 99.8% and the best cut-off value was
448 U/mL. AUROC between severe and non-severe COVID-19 patients using T0 KL-6 concentrations
by AIA360 was 97.4% and the best cut-off value was 398 U/mL. According to T0 KL-6 concentrations
in COVID-19 patients, Bland–Altman difference analysis revealed a mean bias of 78 ± 174.8; while
using T1 KL-6 concentrations in COVID-19 patients, Bland–Altman difference analysis revealed a
mean bias of 48 ± 126 (95% limits of agreement −199–295) between the Lumipulse G600 II and the
AIA360 systems. Conclusions: In conclusion, our study demonstrated that CLEIA and FEIA methods
for serum KL-6 detection are comparable and reliable. KL-6 was confirmed as an easily detectable
and effective biomarker to identify severe COVID-19 patients.

Keywords: Krebs von den Lungen-6; COVID-19; analytical validation

Copyright: © 2022 by the authors.

Licensee MDPI, Basel, Switzerland.
This article is an open access article 1. Introduction
distributed under the terms and
Krebs von den Lungen-6 (KL-6) is a one of the transmembrane mucins in the lung
conditions of the Creative Commons
Attribution (CC BY) license (https://
expressed by injured or regenerating pneumocytes type II [1,2]. In 1992, the diagnostic
creativecommons.org/licenses/by/
division of Eidia Co., Ltd. (Tokyo, Japan) performed a wide-ranging pioneer study on KL-6
4.0/).
as a serum biomarker of lung diseases and these findings led to the development of an

Int. J. Environ. Res. Public Health 2022, 19, 2176. https://doi.org/10.3390/ijerph19042176 https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2022, 19, 2176 2 of 10

enzyme-linked immunosorbent assay (ELISA) to determine the absolute amount of KL-6 in

clinical samples [3]. In 1999, a chemiluminescent enzyme immunoassay (CLEIA) system
became available to detect serum KL-6 levels in ordinary Japanese clinical settings, but not
in Western European countries [4]. At the same time, KL-6 was proposed as diagnostic
marker for fibrotic interstitial lung disease (ILD) and over the years, it demonstrated its
usefulness as a marker for predictive prognosis and response to antifibrotic therapy [5–14].
To date, more than 400 publications are available about the role of KL-6 in ILD patients, es-
pecially on idiopathic pulmonary fibrosis (IPF), as the most common fibrotic ILD. Moreover,
KL-6 has been proposed as bioindicator of acute respiratory distress syndrome (ARDS)
and infective pneumonia [15–19]. Since outbreak of the severe acute respiratory syndrome
coronavirus 2 (SARS-CoV-2) pandemic, KL-6 has also been proposed as a disease severity
marker [20–23]. The pathogenesis of COVID-19 (as the lung disease caused by SARS-CoV-2
is defined) is unclear, though it is postulated that elevated serum concentrations of proin-
flammatory cytokines and oxidative stress mediators contribute to lung injury, facilitating
the onset of an acute respiratory syndrome (similar to ARDS). Host susceptibility and
virus-induced direct cytopathic effects against type I and II pneumocytes are suspected to
play a crucial role in mediating and perpetuating lung damage. Our research group firstly
reported elevated serum concentrations of KL-6 in critical COVID-19 patients and our
results were soon confirmed by other researchers [20,22,23]. We further serially evaluated
serum KL-6 behaviour in a population of COVID-19 hospitalized patients during follow-up,
in order to investigate its potential role in predicting clinical course of disease [21]. All
research articles reported the KL-6 assay detected through CLEIAcleia method by Lu-
mipulse G600/G1200 instrument. In the present study, KL-6 assay was analysed by the
fluorescent enzyme immunoassay (FEIA) method through Tosoh AIA-360 and compared
with analytical results by Lumipulse G600 in a population of COVID-19 patients.

2. Methods
2.1. Study Population
Seventy-five patients (median age, IQR 68 (58–77) years), all hospitalized for COVID-19
interstitial pneumonia from August 2021 to December 2021 at Siena COVID Unit, were
prospectively and consecutively enrolled. Patients who did not give informed consent to
the study or who had a previous diagnosis of interstitial lung diseases, cancer or chronic
obstructive pulmonary diseases were excluded.
Patients were divided into severe and non-severe groups, according to respiratory
impairment and clinical management. All patients in the severe group (median age,
IQR 71 (62–79) years) underwent intubation and mechanical ventilation in the COVID
intensive care unit (ICU), while non-severe patients (not requiring intubation) included
hospitalized subjects requiring pharmacological treatments and oxygen supplementation
or non-invasive ventilation.
Eleven out of 75 COVID-19 hospitalized patients were excluded because they were
affected by concomitant malignancies. Our sixty-four patients underwent serum sampling
specifically for KL-6 assessment on hospital admission (t0).
After hospital discharge (t1) (median IQR, three (2–5) months), twenty-four patients
underwent follow-up evaluations including physical examination, lung function tests,
diffusing capacity of the lung for carbon monoxide (DLCO), blood gas analysis and high-
resolution computed tomography (HRCT) of the chest. CT features (fibrotic interstitial lung
abnormalities, ground glass opacities and air-trapping) were evaluated by on-site radiol-
ogists experienced in interstitial lung diseases. All patients gave their written informed
consent to the study for clinical data collection. The study was approved by our local ethics
committee (C.E.A.V.S.E. Markerlung 17431).

2.2. KL-6 Assay

Assay of serum KL-6 is based on agglutination of sialylated carbohydrate antigen
with KL-6 mAb reagent (Fujirebio Europe, UK and Tosoh Biosciences). Concentrations
Int. J. Environ. Res. Public Health 2022, 19, 2176 3 of 10

(expressed in U/mL) were determined by measuring changes in absorbance as described

in previous papers [10,22,24]. KL-6 was measured by two methods, CLEIA and FEIA, by
Lumipulse G600 II system and AIA 360, respectively. The tests were conducted according
to the manufacturer’s protocol.

2.3. KL-6: Quality Control of Analytical Determinations and Comparison of Results

Serum concentrations of KL-6 were determined with two automatic biochemical
analyzer, Lumipulse G600 II (Fujirebio, Japan) and AIA 360 (Tosoh Biosciences) system that
adopted CLEIA and FEIA method, respectively.
Precision was evaluated according to CLSI EP5-A3 guidelines [25] using two concen-
trations of reagent control serum assays: L1 (Fujirebio, Europe) (mean 339 (range 272–406)
U/mL and L2 (mean 852 (range 682–1022) U/mL. Control serum assays were performed
in duplicate, twice a day for 21 consecutive days. To compare the two instruments, we
measured 49 serum samples from hospitalized COVID-19 patients and nine of them fol-
lowed 3 months from hospital discharge. Comparisons were performed according to CLSI
EP9-A3 guidelines [26].

2.4. Statistical Analysis

The results are reported as means ± SD or as medians and inter quartiles (25th and
75th percentiles) for continuous variables. The Shapiro–Wilk test showed that the data
did not have a normal distribution. To evaluate linearity, the coefficient of determination
(R2) was determined by multiple linear regression. To evaluate precision, the coefficient of
variation was calculated. Bland–Altman analysis was performed for assessing agreement
and bias between systems [27].
One-way ANOVA non-parametric test (Kruskal–Wallis test) and Dunn test were
therefore performed for multiple comparisons. The Chi-square test was applied to cat-
egorical variables. The validity of KL-6 concentrations used to distinguish severe from
non-severe patients was assessed by areas under (AUC) receiver operating characteristic
(ROC) curve. Sensitivity, specificity, and positive and negative predictive values (PPV and
NPV, respectively) were calculated for cut-offs of the different variables. The Youden index
(J = max [sensitivity + specificity − 1]) was used to establish the best cut-offs. Statistical
analysis and graphic representation of the data were performed by GraphPad Prism 9.3 and
XLSTAT 2021 software.

3. Results
3.1. Study Population
Demographic and immunologic data COVID-19 population at T0, including KL-6
concentrations, are reported in Table 1.
Stratifying patients according to severity, 29 of our patients needed intensive care
unit (ICU) and mechanical ventilation. There was a prevalence of males in both groups:
68.9% and 80% in non-severe and severe patients, respectively. Regarding symptoms,
19 out of 35 (54%) non-severe patients and 18 out of 29 (62%) severe patients showed at
least two symptoms at onset (fever and dyspnoea). Twenty-five out of 64 patients were
without comorbidities. In particular, the severe group included 12 patients with arterial
hypertension, four with diabetes and three with heart failure. Two severe patients died
during hospital discharge. Thirteen non-severe patients showed arterial hypertension,
three dyslipidaemia and four heart failure.
Int. J. Environ. Res. Public Health 2022, 19, 2176 4 of 10

Table 1. Demographic and immunological data of COVID-19 hospitalized patients stratified accord-
ing to disease severity. Abbreviation: WBC, white blood cells; NS, not significant; KL-6, Krebs von
den Lungen-6. All data were expressed as median and interquartile range.

Parameters Severe (n = 29) Non-Severe (n = 35) p Value

Age (median IQR) 71 (62–79) 63 (57–73) NS
Gender, M/F 20/9 28/7 NS
T0 KL-6 (U/mL),
827 (599–1103) 293 (220–345) <0.0001
Lumipulse G600II
T0 KL-6 (U/mL),
658 (508–936) 282 (209–346) <0.0001
AIA360
Blood count
Lymphocytes % 13.6 (6.4–15.9) 18.4 (15.9–22.5) 0.041
Neutrophils % 73.5 (67.5–77.1) 77 (68–81) NS
Eosinophils % 0.2 (0–1.4) 0.08 (0.02–0.1) NS
Basophils % 0.15 (0.10–0.17) 0.3 (0.2–0.4) NS
Monocytes % 7.8 (5.38.5) 7.4 (2.9–7.9) NS
WBC (cells/µL) 5.2 (3.7–6.9) 4.9 (3.2–7.1) NS
CRP (mg/L) 5.84 (2.27–6.9) 4.64 (0.64–9.7) 0.037
LDH (U/L) 351 (295–589) 252 (232–330) NS
Glycemia (mg/dL) 104 (105–224) 96 (76–110) NS
Lipase (U/L) 24 (14–28) 20 (17–23) NS
Pancreatic Amylase
30 (18–36) 28 (16–42) NS
(U/L)
AST (U/L) 33 (22–42) 24 (19–39) NS
ALT (U/L) 32 (13–49) 21 (13–29) NS

3.2. KL-6 Assay and Analytical Validation

In the severe group, KL-6 concentrations evaluated by Lumipulse G600II were signifi-
cantly higher than in non-severe patients (p < 0.0001) as well as evaluating through AIA360
(p < 0.0001) (Figure 1).
Int. J. Environ. Res. Public Health 2022, 18, x FOR PEER
There REVIEW
were 5 of
no significant differences of KL-6 concentrations between severe and 11
non-
severe patients at T1 evaluating by Lumipulse G600II (395 (291–912) vs. 244 (188–352),
respectively) and AIA360 (333 (285–472) vs. 254 (277–308), respectively) (p > 0.05).
✱✱✱✱

2000 ✱✱✱✱

1500
KL-6 (U/ml)

1000

500

er
e
er
e re re re re re re
ev ev ve ve ve ve ve ve
-s -s - se - se se se se se
n n 0 1 0 1
n on n on no no T T
0
T
0
T
0 0
T0 T1 T0 T1 60 60 36 36
0 0 0 0 G G IA IA
60 60 36 36 se se A A
G G IA IA ul ul
se se A A ip ip
ul ul um um
ip ip L L
m m
Lu Lu

Figure
Figure 1.
1. KL-6
KL-6 concentrations
concentrations detected
detected by
by Lumipulse
Lumipulse G600II
G600II and
and AIA360
AIA360 at
at T0
T0 and
and T1 stratifying
stratifying
COVID-19 patients according to disease severity. **** p < 0.0001.
COVID-19 patients according to disease severity. **** p < 0.0001.

Receiver operating curve (ROC) analysis showed that KL-6 concentrations, evaluated
by Lumipuse G600II (Figure 2), distinguished severe from non-severe COVID-19 patients
with an area under the curve (AUC) of 99.8% and the best cut-off value was 448 U/mL
with 97.1% specificity and 100% sensitivity.
 n- n- n- n- T0 T1 T0 T1
no no no no 0 0 0 0
T0 T1 T0 T1 60 60 36 36
0 0 0 0 G G IA IA
60 60 36 36 se se A A
G G IA IA ul ul
ls
e
ls
e A A ip ip
m m
i pu i pu Lu Lu
m m
Lu Lu

Int. J. Environ. Res. Public Health 2022,Figure

19, 21761. KL-6 concentrations detected by Lumipulse G600II and AIA360 at T0 and T1 stratifying
5 of 10
COVID-19 patients according to disease severity. **** p < 0.0001.

Receiver operating
Receiver operating curve
curve (ROC)
(ROC) analysis
analysis showed
showed that
that KL-6
KL-6 concentrations,
concentrations, evaluated
evaluated
by Lumipuse G600II (Figure 2), distinguished severe from non-severe
by Lumipuse G600II (Figure 2), distinguished severe from non-severe COVID-19COVID-19 patients
patients
with an area under the curve (AUC) of 99.8% and the best cut-off value was
with an area under the curve (AUC) of 99.8% and the best cut-off value was 448 U/mL 448 U/mL
with 97.1% specificity and 100% sensitivity.
with 97.1% specificity and 100% sensitivity.

(a) (b)
Figure 2.
Figure 2. (a)
(a)Area
Areaunder
underthe receiver
the operating
receiver curve
operating (ROC)
curve analysis
(ROC) distinguishing
analysis severe
distinguishing and non-
severe and
severe patients according to T0 KL-6 concentrations detected by Lumipulse G600II. (b) The percent-
non-severe patients according to T0 KL-6 concentrations detected by Lumipulse G600II. (b) The
ages of the false negative, true negative, false positive and true positive patients for the KL-6 con-
percentages of the false negative, true negative, false positive and true positive patients for the KL-6
centrations at T0 detected by Fujirebio reagent.
concentrations at T0 detected by Fujirebio reagent.
AUROC between severe and non-severe COVID-19 patients using T0 KL-6 concen-
AUROC between severe and non-severe COVID-19 patients using T0 KL-6 concen-
Int. J. Environ. Res. Public Health 2022, trations
18, x FOR evaluated
PEER REVIEW 6 of 11
through
AIA360 (Figure 3) was 97.4% and the best cut-off value was
trations evaluated through AIA360 (Figure 3) was 97.4% and the best cut-off value was
398 U/mL with 89% specificity and 97% sensitivity.
398 U/mL with 89% specificity and 97% sensitivity.

(a) (b)
Figure
Figure 3.
3.(a)
(a)Area
Areaunder
underthethe
receiver operating
receiver curve
operating (ROC)
curve analysis
(ROC) distinguishing
analysis severe
distinguishing and non-
severe and
severe patients according to T0 KL-6 concentrations detected by AIA360. (b) The percentages of the
non-severe patients according to T0 KL-6 concentrations detected by AIA360. (b) The percentages of
false negative, true negative, false positive and true positive patients for the KL-6 concentrations at
the false negative, true negative, false positive and true positive patients for the KL-6 concentrations
T0 detected by Tosoh biosciences reagent.
at T0 detected by Tosoh biosciences reagent.
The precision of KL-6 analytical determinations (CV%) for L1 and L2 reagent concen-
trations (Figure 4a) was 1.84% for the lower and 1.47% for the higher, respectively. Using
L1 and L2 reagent concentrations measured by AIA360, Bland–Altman difference analysis
(Figure 4b) revealed a mean bias of −15 ± 2.39 (95% limits of agreement −20–−10).

200
L1
 (a)  (b) 
Figure 3. (a) Area under the receiver operating curve (ROC) analysis distinguishing severe and non‐
severe patients according to T0 KL‐6 concentrations detected by AIA360. (b) The percentages of the 
false negative, true negative, false positive and true positive patients for the KL‐6 concentrations at 
Int. J. Environ. Res. Public Health 2022, 19, 2176 6 of 10
T0 detected by Tosoh biosciences reagent. 

The  precision  of  KL‐6  analytical  determinations  (CV%)  for  L1  and  L2  reagent 
The precision of KL-6 analytical determinations (CV%) for L1 and L2 reagent concen-
concentrations (Figure 4a) was 1.84% for the lower and 1.47% for the higher, respectively. 
trations (Figure 4a) was 1.84% for the lower and 1.47% for the higher, respectively. Using
Using L1 and L2 reagent concentrations measured by AIA360, Bland–Altman difference 
L1 and L2 reagent concentrations measured by AIA360, Bland–Altman difference analysis
analysis (Figure 4b) revealed a mean bias of −15 ± 2.39 (95% limits of agreement −20–−10). 
(Figure 4b) revealed a mean bias of −15 ± 2.39 (95% limits of agreement −20–−10).

200
L1
190 L2

180

170

160

150
1

10

11

12

13

14

15

16

17

18

19

20

21
ay

ay

ay

ay

ay

ay

y
Da

Da

Da

ay

ay

ay

ay

ay

ay

ay

ay

ay

ay

ay

ay
D

D
 
(a) 

 
(b) 

Figure 4. (a) the linearity of L1 and L2 reagent concentrations by Fujirebio analysed on AIA360
instrument. (b) Bland–Altman analysis for L1 and L2 concentrations for assessing agreement and
Bias between systems.

  The QQ plot (Figure S1) was obtained using four degrees of freedom (Lumipulse KL-6
T0 and KL-6 T1, AIA360 KL-6 T0 and KL-6 T1) with the coefficient of determination (R2) of
the regression analysis of 0.5977.
According to T0 KL-6 concentrations in COVID-19 patients, Bland–Altman differ-
ence analysis (Figure 5a) revealed a mean bias of 78 ± 174.8 (95% limits of agreement
−263.7–421.6) between the Lumipulse G600 II and the AIA360 systems. Figure 5b indi-
cates the regression line between the two systems; correlation coefficient between the two
methods is r = −0.661 (95% confidence interval, CI = −0.78, −0.50, p = 0.001); that could be
evaluated as a very good agreement.
According to T1 KL-6 concentrations in COVID-19 patients, Bland–Altman difference
analysis (Figure 6a) revealed a mean bias of 48 ± 126 (95% limits of agreement −199–295)
between the Lumipulse G600 II and the AIA360 systems. Figure 6b indicates the regression
line between the two systems; correlation coefficient between the two methods is r = −0.642
(95% confidence interval, CI = −0.83, −0.32, p = 0.045).
 as a very good agreement.
According to T0 KL-6 concentrations in COVID-19 patients, Bland–Altman difference
analysis (Figure 5a) revealed a mean bias of 78 ± 174.8 (95% limits of agreement −263.7—
421.6) between the Lumipulse G600 II and the AIA360 systems. Figure 5b indicates the
regression line between the two systems; correlation coefficient between the two methods
Int. J. Environ. Res. Public Health 2022,is = −0.661 (95% confidence interval, CI = −0.78, −0.50, p = 0.001); that could be evaluated
19,r 2176 7 of 10
as a very good agreement.

(a) (b)
Figure 5. (a) Bland–Altman analysis using T0 KL-6 concentrations detected by two methods, CLEIA
and FEIA by Lumipulse G600II and AIA360 system, respectively, for assessing agreement and Bias
between systems. (b) Scatter plot analysis to perform linear correlation between KL-6 T0 Lumipulse
G600II and AIA360.

(a) According to T1 KL-6 concentrations in COVID-19 patients, (b) Bland–Altman difference

analysis
Figure 5. (Figure
5. (a) 6a) revealed a mean bias of 48 ± 126 (95% limits of by
agreement −199–295)
Figure (a) Bland–Altman
Bland–Altman analysis using T0
analysis using T0 KL-6
KL-6 concentrations detected
concentrations detected by two
two methods, CLEIA
methods, CLEIA
between
and FEIA the Lumipulse
by Lumipulse G600and
G600II II and the system,
AIA360 AIA360respectively,
systems. Figure 6b indicates
for assessing the and
agreement regres-
Bias
and FEIA by Lumipulse G600II and AIA360 system, respectively, for assessing agreement and Bias
sion linesystems.
between between(b)the two plot
Scatter systems; correlation
analysis to performcoefficient between
linear correlation the two
between methods
KL-6 is r =
T0 Lumipulse
between systems. (b) Scatter plot analysis to perform linear correlation between KL-6 T0 Lumipulse
−0.642 confidence interval, CI = −0.83, −0.32, p = 0.045).
(95%AIA360.
G600II and
G600II and AIA360.
According to T1 KL-6 concentrations in COVID-19 patients, Bland–Altman difference
analysis (Figure 6a) revealed a mean bias of 48 ± 126 (95% limits of agreement −199–295)
between the Lumipulse G600 II and the AIA360 systems. Figure 6b indicates the regres-
sion line between the two systems; correlation coefficient between the two methods is r =
−0.642 (95% confidence interval, CI = −0.83, −0.32, p = 0.045).

(a) (b)
Figure
Figure 6.
6. (a)
(a) Bland–Altman
Bland–Altman analysis
analysis using
using T1
T1 KL-6
KL-6 concentrations
concentrations detected
detected by
by two
two methods,
methods, CLEIA
CLEIA
and FEIA by Lumipulse G600II and AIA360 system, respectively, for assessing agreement and Bias
and FEIA by Lumipulse G600II and AIA360 system, respectively, for assessing agreement and Bias
between systems. (b) Scatter plot analysis to perform linear correlation between KL-6 T1 Lumipulse
between systems. (b) Scatter plot analysis to perform linear correlation between KL-6 T1 Lumipulse
G600II and AIA360.
G600II and AIA360.
(a) (b)
4. Discussion
Figure
In6.the
(a) present
Bland–Altman
study,analysis
the KL-6using T1 KL-6 concentrations
concentrations in COVID-19detected by twoby
patients methods, CLEIA
two different
and FEIA by Lumipulse G600II and AIA360 system, respectively, for assessing agreement
methods, CLEIA and FEIA by Lumipulse G600II (Fujirebio, Europe, Ghent, Belgium) and and Bias
between systems. (b) Scatter plot analysis to perform linear correlation between KL-6 T1 Lumipulse
AIA360 (Tosoh, Biosciences, Tokyo, Japan) system, were evaluated.
G600II and AIA360.
Despite the large number of performed studies, no literature is available about the
quantification of serum KL-6 through FEIA assay by AIA360. Most articles evaluated KL-6
through CLEIA and enzyme-linked immunosorbent assay (ELISA) assays. In 1992, the
diagnostic division of Eidia Co., Ltd. (Tokyo, Japan) performed a wide-ranging pioneer
study on KL-6 as a serum biomarker of lung diseases and these findings led to the develop-
ment of an ELISA that enabled determination of the absolute amount of KL-6 in clinical
samples [3]. In 1999, the Japanese Health Insurance Program approved KL-6 as a diagnostic
marker of ILD. The CLEIA system became available to detect serum KL-6 levels in ordinary
Japanese clinical settings, but not in Western European countries, including Italy. The assay
Int. J. Environ. Res. Public Health 2022, 19, 2176 8 of 10

takes only 1 h to perform through the CLEIA method by Lumipulse G600II and the first
result is available in 35 min.
FEIA was adopted for the first time in the present preliminary study to evaluate KL-6
concentrations in serum samples from COVID-19 hospitalized patients including a cohort
of patients followed after hospital discharge in order to assess the comparison between two
methods: CLEIA and FEIA. The usefulness to have another method to perform KL-6 assay
in serum samples is extremely important, especially if it takes into account the advantages
of the FEIA method through AIA360 instrument in terms of time consumption (36 tests per
hour, first result ~ 20 min) and continuous processing.
KL-6 has various pathophysiologic roles (inhibiting cell-cell adhesion of epithelial
cells [28] including fibroblast migration [29] and several studies indicated that after bronchial
epithelium damage and reparative proliferation processes, KL-6 is released from the re-
generated type II pneumocytes into the blood [2,30–33]. The first analysis of serum KL-6
concentrations was performed by enzyme-linked immunosorbent assay (ELISA) dating to
1992 [3]. Over the years, a CLEIA system [4], that takes only 1 h to be performed, became
available to detect serum KL-6 levels in ordinary Japanese clinical settings, but not in
Western European countries.
Increased KL-6 concentrations have been previously reported in different pulmonary
diseases, including hypersensitivity pneumonitis [34], idiopathic pulmonary fibrosis [13,14],
acute respiratory distress syndrome [15], lung cancer [35], and pulmonary sarcoidosis [8].
Moreover, KL-6 concentrations reflecting the extent of damage and regeneration of type II
pneumocytes can predict the risk of illness or death in subjects suffering from lung diseases
associated with rheumatologic disorders [11]. A cut-off value of 465 U/mL was recently
established by Lumipulse G600II system to distinguish fibrotic ILD patients from healthy
subjects and those patients with other non-fibrotic lung diseases.
Since the outbreak of the SARS-CoV-2 pandemic, our group of researchers suggested
for the first time the role of KL-6, detected through the CLEIA method, as a disease severity
biomarker for COVID-19 patients and contributed to the definition of the natural course of
COVID-19 as the normalization of peripheral KL-6 concentrations during follow-up [21,22].
These data were confirmed in the present study evaluating 64 COVID-19 patients hospital-
ized between August 2021 to December 2021 at Siena COVID Unit University Hospital by
comparing for the first time two different methods, CLEIA and FEIA by Lumipulse G600II
and AIA360 system, respectively. The analytical performance of KL-6 assay kit, based on
CLEIA and FEIA methods through two different instruments, was confirmed to be reliable:
cut-off values discriminating severe from non-severe hospitalized COVID-19 patients were
448 and 398 U/mL for Lumipulse G600 II and AIA360, respectively.
It can be speculated that KL-6 has a crucial role as a key molecule involved in epithelial-
mesenchymal interactions, though the pathogenetic pathways involved in serum KL-6
increase are unknown as well as KL-6’s prognostic and predictive role as biomarker in
COVID-19 patients.
In conclusion, our study demonstrated that CLEIA and FEIA methods for serum
KL-6 detection are comparable and reliable. KL-6 was confirmed an easily detectable and
effective biomarker to identify severe COVID-19 patients.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/ijerph19042176/s1, Figure S1: The QQ plot obtained using four
degrees of freedom (Lumipulse KL-6 T0 and KL-6 T1, AIA360 KL-6 T0 and KL-6 T1) with the
coefficient of determination (R2) of the regression analysis of 0.5977.
Author Contributions: Conceptualization, M.d., L.B.; methodology, M.d., D.C., S.G., L.B.; software,
M.d.; validation, M.d., P.C., E.C., B.F., F.D., E.B., L.B.; writing—original draft preparation, M.d., P.C.,
S.G., D.C., E.C., B.F., F.D., E.B., L.B. All authors have read and agreed to the published version of
the manuscript.
Funding: This research project is not funded.
Int. J. Environ. Res. Public Health 2022, 19, 2176 9 of 10

Institutional Review Board Statement: The study was conducted according to the guidelines of the
Declaration of Helsinki and approved by the Ethics Committee of Comitato Etico Area Vasta Sud Est,
Tuscany (Markerlung, 17431).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the
study. Written informed consent has been obtained from the patient(s) to publish this paper.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Acknowledgments: The authors acknowledged Tosoh Biosciences for providing reagents for KL-6 anal-
ysis by AIA360 instrument. Siena COVID Unit: Francesca Montagnani (francesca.montagnani@unisi.it),
Anna Perrone (A.perrone@ao-siena.toscana.it), Federico Franchi (federico.franchi@dbm.unisi.it), Sabino
Scolletta (sabino.scolletta@dbm.unisi.it), Serafina Valente (serafina.valente@ao-siena.toscana.it), Lucia
Migliorini (l.migliorini@ao-siena.toscana.it), Barbara Rossetti (brossetti1982@gmail.com), Cecilia Va-
gaggini (cecivag@gmail.com), Pier Leopoldo Capecchi (pierleopoldo.capecchi@unisi.it), Maria Grazia
Cusi (mariagrazia.cusi@unisi.it), Egidio Mastrocinque (egidio-mastrocinque@alice.it), Matteo Cameli
(matteo.cameli@unisi.it), Marco Antonio Bellini (acropo-lis@ao-siena.toscana.it), Nicola De Stefano
(nicola.destefano@unisi.it).
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations

KL-6: Krebs von den Lungen-6; ILD, interstitial lung diseases; COVID-19, coronavirus
disease-19; CRP, C-reactive protein.

References
1. D’Alessandro, M.; Bergantini, L.; Cameli, P.; Vietri, L.; Lanzarone, N.; Alonzi, V.; Pieroni, M.; M Refini, R.; Sestini, P.;
Bonella, F.; et al. Krebs Von Den Lungen-6 as a Biomarker for Disease Severity assessment in Interstitial Lung Disease: A
Comprehensive Review. Biomark. Med. 2020, 14, 665–674. [CrossRef] [PubMed]
2. Hu, Y.; Wang, L.-S.; Jin, Y.-P.; Du, S.-S.; Du, Y.-K.; He, X.; Weng, D.; Zhou, Y.; Li, Q.-H.; Shen, L.; et al. Serum Krebs Von Den
Lungen-6 Level as a Diagnostic Biomarker for Interstitial Lung Disease In Chinese Patients. Clin. Respir. J. 2017, 11, 337–345.
[CrossRef] [PubMed]
3. Hamada, H.; Kohno, N.; Akiyama, M.; Hiwada, K. Monitoring of Serum Kl-6 Antigen in A Patient with Radiation Pneumonia.
Chest 1992, 101, 858–860. [CrossRef] [PubMed]
4. Yokoyama, A.; Kohno, N.; Kondo, K.; Ueda, S.; Hirasawa, Y.; Watanabe, K.; Takada, Y.; Hiwada, K. Comparative Evaluation of
Sialylated Carbohydrate Antigens, Kl-6, Ca19-9 and Slx as Serum Markers for Interstitial Pneumonia. Respirology 1998, 3, 199–202.
[CrossRef] [PubMed]
5. Yokoyama, A.; Kondo, K.; Nakajima, M.; Matsushima, T.; Takahashi, T.; Nishimura, M.; Bando, M.; Sugiyama, Y.; Totani, Y.;
Ishizaki, T.; et al. Prognostic Value of Circulating Kl-6 In Idiopathic Pulmonary Fibrosis. Respirology 2006, 11, 164–168. [CrossRef]
6. Ishikawa, N.; Hattori, N.; Yokoyama, A.; Kohno, N. Utility of Kl-6/Muc1 In The Clinical Management of Interstitial Lung
Diseases. Respir. Investig. 2012, 50, 3–13. [CrossRef]
7. Iwata, Y.; Wada, T.; Furuichi, K.; Kitagawa, K.; Kokubo, S.; Kobayashi, M.; Sakai, N.; Yoshimoto, K.; Shimizu, M.;
Kobayashi, K.; et al. Serum Levels of Kl-6 Reflect Disease Activity of Interstitial Pneumonia associated with Anca-Related
Vasculitis. Intern. Med. 2001, 40, 1093–1097. [CrossRef]
8. Bergantini, L.; Bianchi, F.; Cameli, P.; Mazzei, M.A.; Fui, A.; Sestini, P.; Rottoli, P.; Bargagli, E. Prognostic Biomarkers of Sarcoidosis:
A Comparative Study of Serum Chitotriosidase, Ace, Lysozyme, and Kl-6. Dis. Markers 2019, 2019, 8565423. [CrossRef]
9. Bergantini, L.; Bargagli, E.; Cameli, P.; Cekorja, B.; Lanzarone, N.; Pianigiani, L.; Vietri, L.; Bennett, D.; Sestini, P.; Rottoli, P.
Serial Kl-6 Analysis In Patients with Idiopathic Pulmonary Fibrosis Treated with Nintedanib. Respir. Investig. 2019, 57, 290–291.
[CrossRef]
10. D’alessandro, M.; Bergantini, L.; Cameli, P.; Lanzarone, N.; Antonietta Mazzei, M.; Alonzi, V.; Sestini, P.; Bargagli, E. Serum Kl-6
Levels in Pulmonary Langerhans’ Cell Histiocytosis. Eur. J. Clin. Investig. 2020, 50, e13242. [CrossRef]
11. D’alessandro, M.; Bellisai, F.; Bergantini, L.; Cameli, P.; D’alessandro, R.; Mazzei, M.A.; Gentili, F.; Conticini, E.; Selvi, E.;
Frediani, B.; et al. Prognostic Role of Kl-6 in Ssc-Ild Patients with Pleuroparenchymal Fibroelastosis. Eur. J. Clin. Investig. 2021, 51,
E13543. [CrossRef] [PubMed]
12. D’alessandro, M.; Bergantini, L.; Cameli, P.; Perillo, F.; Remediani, L.; Refini, R.M.; Pieroni, M.; Mazzei, M.A.; Sestini, P.;
Bargagli, E. Prognostic Role of Kl-6 In Lymphangioleiomyomatosis Patients. Minerva Med. 2021. [CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2022, 19, 2176 10 of 10

13. D’alessandro, M.; Bergantini, L.; Cameli, P.; Pieroni, M.; Refini, R.M.; Sestini, P.; Bargagli, E. Serum Concentrations of Kl-6 In
Patients with Ipf and Lung Cancer and Serial Measurements of Kl-6 In Ipf Patients Treated with Antifibrotic Therapy. Cancers
2021, 13, 689. [CrossRef] [PubMed]
14. D’alessandro, M.; Bergantini, L.; Torricelli, E.; Cameli, P.; Lavorini, F.; Pieroni, M.; Refini, R.M.; Sestini, P.; Bargagli, E. Systematic
Review and Metanalysis of Oncomarkers in Ipf Patients and Serial Changes of Oncomarkers in A Prospective Italian Real-Life
Case Series. Cancers 2021, 13, 539. [CrossRef]
15. Sato, H.; Callister, M.E.J.; Mumby, S.; Quinlan, G.J.; Welsh, K.I.; Dubois, R.M.; Evans, T.W. Kl-6 Levels Are Elevated in Plasma
From Patients with Acute Respiratory Distress Syndrome. Eur. Respir. J. 2004, 23, 142–145. [CrossRef]
16. Nakamura, H.; Tateyama, M.; Tasato, D.; Haranaga, S.; Yara, S.; Higa, F.; Ohtsuki, Y.; Fujita, J. Clinical Utility of Serum
Beta-D-Glucan and Kl-6 Levels in Pneumocystis Jirovecii Pneumonia. Intern. Med. 2009, 48, 195–202. [CrossRef]
17. Urabe, N.; Sakamoto, S.; Sano, G.; Ito, A.; Sekiguchi, R.; Homma, S. Serial Change In Serum Biomarkers during Treatment of
Non-Hiv Pneumocystis Pneumonia. J. Infect. Chemother. 2019, 25, 936–942. [CrossRef] [PubMed]
18. Arai, Y.; Obinata, K.; Sato, Y.; Hisata, K.; Tadokoro, R.; Tawa, T.; Kinoshita, K. Clinical Significance of the Serum Surfactant Protein
D and Kl-6 Levels in Patients with Measles Complicated By Interstitial Pneumonia. Eur. J. Pediatr. 2001, 160, 425–429. [CrossRef]
19. Kawasaki, Y.; Aoyagi, Y.; Abe, Y.; Go, H.; Imamura, T.; Kaneko, M.; Ito, M.; Katayose, M.; Hashimoto, K.; Hosoya, M. Serum Kl-6
Levels as a Biomarker of Lung Injury in Respiratory Syncytial Virus Bronchiolitis. J. Med. Virol. 2009, 81, 2104–2108. [CrossRef]
20. Bergantini, L.; Bargagli, E.; D’alessandro, M.; Refini, R.M.; Cameli, P.; Galasso, L.; Scapellato, C.; Montagnani, F.; Scolletta, S.;
Franchi, F.; et al. Prognostic Bioindicators in Severe COVID-19 Patients. Cytokine 2021, 141, 155455. [CrossRef]
21. D’alessandro, M.; Bergantini, L.; Cameli, P.; Curatola, G.; Remediani, L.; Bennett, D.; Bianchi, F.; Perillo, F.; Volterrani, L.;
Mazzei, M.A.; et al. Serial Kl-6 Measurements in COVID-19 Patients. Intern. Emerg. Med. 2021, 16, 1541–1545. [CrossRef]
[PubMed]
22. D’alessandro, M.; Cameli, P.; Refini, R.M.; Bergantini, L.; Alonzi, V.; Lanzarone, N.; Bennett, D.; Rana, G.D.; Montagnani, F.;
Scolletta, S.; et al. Serum Kl-6 Concentrations as a Novel Biomarker of Severe COVID-19. J. Med. Virol. 2020, 92, 2216–2220.
[CrossRef] [PubMed]
23. D’alessandro, M.; Cameli, P.; Bergantini, L.; Franchi, F.; Scolletta, S.; Bargagli, E. Serum Concentrations of Krebs Von Den Lungen-6
in Different COVID-19 Phenotypes. J. Med. Virol. 2020, 93, 657. [CrossRef] [PubMed]
24. Awano, N.; Inomata, M.; Kuse, N.; Tone, M.; Takada, K.; Muto, Y.; Fujimoto, K.; Akagi, Y.; Mawatari, M.; Ueda, A.; et al. Serum
Kl-6 Level Is A Useful Biomarker For Evaluating the Severity of Coronavirus Disease 2019. Respir. Investig. 2020, 58, 440–447.
[CrossRef]
25. Ep05a3: Evaluating Quantitative Measurement Precision. Available online: Https://Clsi.Org/Standards/Products/Method-
Evaluation/Documents/Ep05/ (accessed on 26 December 2021).
26. Budd, J.R.; Durham, A.P.; Gwise, T.E.; Hawkins, D.M.; Holland, M.; Iriarte, B.; Kallner, A.; Linnet, K.; Magari, R.; Vaks, J.E.
Measurement Procedure Comparison and Bias Estimation Using Patient Samples; Clinical Laboratory Standards Institute: Wayne, NY,
USA, 2013.
27. Bland, J.M.; Altman, D.G. Statistical Methods For assessing Agreement Between Two Methods of Clinical Measurement. Lancet
1986, 1, 307–310. [CrossRef]
28. Hilkens, J.; Ligtenberg, M.J.; Vos, H.L.; Litvinov, S.V. Cell Membrane-associated Mucins and Their Adhesion-Modulating Property.
Trends Biochem. Sci. 1992, 17, 359–363. [CrossRef]
29. Hirasawa, Y.; Kohno, N.; Yokoyama, A.; Inoue, Y.; Abe, M.; Hiwada, K. Kl-6, A Human Muc1 Mucin, Is Chemotactic For Human
Fibroblasts. Am. J. Respir. Cell Mol. Biol. 1997, 17, 501–507. [CrossRef]
30. Bonella, F.; Costabel, U. Biomarkers In Connective Tissue Disease-associated Interstitial Lung Disease. Semin. Respir. Crit. Care
Med. 2014, 35, 181–200. [CrossRef]
31. Bonella, F.; Volpe, A.; Caramaschi, P.; Nava, C.; Ferrari, P.; Schenk, K.; Ohshimo, S.; Costabel, U.; Ferrari, M. Surfactant Protein D
and Kl-6 Serum Levels in Systemic Sclerosis: Correlation with Lung and Systemic Involvement. Sarcoidosis Vasc. Diffus. Lung Dis.
2011, 28, 27–33.
32. Zhong, D.; Wu, C.; Bai, J.; Hu, C.; Xu, D.; Wang, Q.; Zeng, X. Comparative Diagnostic Efficacy of Serum Krebs Von Den Lungen-6
and Surfactant D For Connective Tissue Disease-associated Interstitial Lung Diseases: A Meta-Analysis. Medicine 2020, 99, E19695.
[CrossRef]
33. Xue, M.; Guo, Z.; Cai, C.; Sun, B.; Wang, H. Evaluation of the Diagnostic Efficacies of Serological Markers Kl-6, Sp-A, Sp-D, Ccl2,
and Cxcl13 In Idiopathic Interstitial Pneumonia. Respiration 2019, 98, 534–545. [CrossRef] [PubMed]
34. Lanzarone, N.; Gentili, F.; Alonzi, V.; Bergantini, L.; D’alessandro, M.; Rottoli, P.; Refini, R.M.; Pieroni, M.; Vietri, L.; Bianchi, F.; et al.
Bronchoalveolar Lavage and Serum Kl-6 Concentrations in Chronic Hypersensitivity Pneumonitis: Correlations with Radiological
and Immunological Features. Intern. Emerg. Med. 2020, 15, 1247–1254. [CrossRef] [PubMed]
35. Miyazaki, K.; Kurishima, K.; Kagohashi, K.; Kawaguchi, M.; Ishikawa, H.; Satoh, H.; Hizawa, N. Serum Kl-6 Levels In Lung
Cancer Patients with or Without Interstitial Lung Disease. J. Clin. Lab. Anal. 2010, 24, 295–299. [CrossRef] [PubMed]